| 摘要: |
| 随着交通、通讯设施的日益完善与经济
的快速发展，城市间各类要素流动更为频繁并
形成城市网络，促进城市动态“流”数据的分析
成为区域空间结构研究新范式。相比单一要素
流，多维要素流可以从更加综合的视角刻画城
市群内部网络联系，识别城市群空间结构特征。
本文通过集成百度迁徙、快递物流线路、百度指
数、企业总部—分支、科技论文合作等多元地
理流数据建立成渝城市群人流、物流、信息流、
资金流、技术流及综合流网络，借助社会网络
分析方法识别网络节点特征并结合位序—规模
法则评估城市体系规模结构，利用核密度分析
法识别多维要素流动主要廊道，结合优势流和
DBSCAN聚类分析成渝城市群空间组团特征。
结果表明：第一，在多维要素流网络中，各节点
层级分化明显，成都市、重庆市是成渝城市群的
两大核心，对多维要素流的集聚扩散能力突出，而其他城市普遍发育不足。第二，重庆市—成都市关联区间联系强度最高，成都市、重庆市与14个
地级市组成的关联区间次之，14个地级市之间组成的关联区间最低，成渝发展主轴、成德绵乐城
市带是要素流动的主要廊道。第三，在优势流约束下，成渝城市群内部形成成都—德阳—眉山、重
庆—广安、南充—遂宁、内江—自贡—宜宾—泸州、乐山—雅安共5个空间聚类，其中南充—遂宁、
内江—自贡—宜宾—泸州具备培育都市圈的潜力。结合本文分析结果和现有规划，建议将多维要
素流网络中心度相对较高的绵阳、南充、宜宾作为次级中心城市培育，在重点发展成都都市圈、重
庆都市圈的同时着力培育南充—遂宁、内自宜泸两大都市圈，促进绵阳市、雅安市、乐山市、达州市
等圈群空隙城市差异化、特色化发展，强化宜宾—泸州—重庆沿江发展轴，逐步优化成渝城市群
空间结构，形成区域协调发展新格局。 |
| 关键词: 空间结构 多维要素流 社会网络分析 位序—规模 核密度估计 DBSCAN聚类 成渝
城市群 |
| DOI：10.13791/j.cnki.hsfwest.20240303 |
| 分类号: |
| 基金项目:国家自然科学基金项目（52078115）；四川省软科学
研究计划项目（2023JDR0094） |
|
| Spatial structure characteristics of Chengdu-Chongqing urban agglomeration from theperspective of multi-dimensional factor flows |
|
ZHANG Yang,LI Juan,WANG Xingping
|
| Abstract: |
| With the increasingly perfect transportation and communication facilities and rapid
economic development, various factor flows more frequently between cities and form a city network,
promoting the analysis of urban dynamic "flow" data to become a new paradigm for regional spatial
structure research. Compared with single factor flow, multi-dimensional factor flows can depict the
network connection and identify the spatial structure characteristics of the urban agglomeration from
a more comprehensive perspective. This paper establishes networks of people flow, logistics flow,
information flow, capital flow, technology flow, and comprehensive flow in the Chengdu-Chongqing
urban agglomeration by integrating multi-dimensional geographic flow data such as Baidu Migration,
express logistics routes, Baidu Index, headquarters-branches of enterprises, and paper cooperation.
By using social network analysis to identify network node characteristics and combining rank-size
rules to evaluate the scale structure of the urban system, and using kernel density analysis method
to identify the main corridors of multi-dimensional factor flows, and combining advantage flow and
density-based spatial Clustering of application with noise clustering analysis to analyze the spatial
clustering characteristics of the Chengdu-Chongqing urban agglomeration. The results show that: In
the multi-dimensional factor flow network, the hierarchical differentiation of each node is obvious.
Chengdu and Chongqing are the central node cities, with prominent agglomeration and diffusion
capabilities of multi-dimensional factor flow, while other cities are generally under developed. The
interval between Chongqing and Chengdu has the strongest connection strength, followed by the
interval of Chengdu, Chongqing, and 14 prefect ure-level cities, and the interval of 14 prefecturelevel
cities is the lowest. The main corridors of factor flow are the Chengdu-Chongqing development
axis and the Chengdu-Deyang-Mianyang-Leshan urban belt. Under the constraint of the advantage
flow, five spatial clusters are formed within the Chengdu-Chongqing urban agglomeration, including
Chengdu-Deyang-Meishan, Chongqing-Guangan, Nanchong-Suining, Neijiang-Zigong-Yibin-
Luzhou, and Leshan-Ya’an. Among them, Nanchong-Suining and Neijiang-Zigong-Yibin-Luzhou
have the potential to develop into metropolitan areas. Based on the analysis results and existing
plans, this paper suggests that Mianyang, Nanchong, and Yibin, which have relatively high centralityin the multi-dimensional element flow network, be cultivated as secondary central cities. While focusing on the development of the Cheng du metropolitan
area and the Chongqing metropolitan area, efforts should be made to cultivate the two metropolitan areas of Nanchong-Suining and Neijiang-Zigong-Yibin-
Luzhou, promote differentiated and characteristic development of urban gaps such as Mian yang, Ya’an, Leshan, and Dazhou, strengthen the development
axis of Yibin-Luzhou-Chongqing along the Yangtze River, and gradually optimize the spatial structure of the Chengdu-Chongqing urban agglomeration,
forming a new pattern of coordinated regional development. |
| Key words: spatial structure multi-dimensional factor flows social network estimation rank-size rule kernel density analysis, density-based spatial
clustering of application with noise (DBSCAN) Chengdu-Chongqing urban agglomeration |
